Home / Performance Metrics / CAGR (Compound Annual Growth Rate)
Return metricBeginner

CAGR (Compound Annual Growth Rate)

CAGR (compound annual growth rate) is the single constant yearly rate that, compounded over the full test period, would take the starting equity to the ending equity, expressed as (End ÷ Start) raised to the power (1 ÷ Years) minus one.

Quick answer: CAGR (compound annual growth rate) is the single constant yearly rate that, compounded over the full test period, would take the starting equity to the ending equity, expressed as (End ÷ Start) raised to the power (1 ÷ Years) minus one.

In simple words

CAGR answers a simple question: at what steady yearly rate did your account grow, ignoring the bumps along the way. It smooths a jagged equity curve into one clean percentage as if the money had grown at the same pace every single year. It is the number you compare across strategies and against a Nifty benchmark, but it says nothing about how rough the ride was.

Purpose

CAGR exists to make growth over different lengths of time comparable: a strategy tested over 3 years and one tested over 8 years cannot be compared by total return, but their CAGRs are on the same annual footing.

Professional explanation

What CAGR actually measures

CAGR is a geometric mean of annual growth, not an arithmetic average of yearly returns. It captures compounding: a year of plus 50 percent followed by a year of minus 50 percent leaves you down 25 percent overall, and CAGR correctly reports a negative annual rate, whereas naively averaging plus 50 and minus 50 would wrongly report zero. Because it is derived only from the first and last equity values and the elapsed time, CAGR is completely blind to the path taken between them.

Why the exponent is one over the number of years

Growing at rate g for Y years multiplies capital by (1 + g) to the power Y. To invert that and recover g from the observed multiple (End ÷ Start), you raise the multiple to the power (1 ÷ Y) and subtract one. The number of years must be measured precisely, including fractions: using 3 instead of 3.25 years inflates the reported CAGR, and the error grows as the period shortens.

Fractional years and trading-day conventions

For a backtest that does not start and end exactly on year boundaries, compute years as calendar days divided by 365.25, or as trading days divided by roughly 252 if you are working on a trading-day clock. Consistency matters more than the exact convention: mixing a 252-day annualiser in one place with a 365-day one in another produces subtly wrong and non-comparable numbers across a research library.

CAGR versus arithmetic mean return

The arithmetic mean of periodic returns always equals or exceeds the geometric (CAGR-implied) mean, and the gap widens with volatility. This volatility drag is real money: two strategies with the same average return but different volatility will have different CAGRs, the calmer one ending richer. This is precisely why CAGR, being geometric, is the honest growth number and arithmetic averages flatter volatile strategies.

The blind spot that makes CAGR dangerous alone

Because CAGR only sees endpoints, two strategies with identical CAGR can have wildly different risk: one may have crept up steadily while the other tripled, collapsed 70 percent, and clawed back. CAGR would rate them equal. This is why CAGR is never reported alone in serious research; it is paired with a drawdown or a risk-adjusted ratio such as Calmar (CAGR divided by max drawdown) so the reader sees both the reward and the pain that produced it.

Formula

CAGR = (End ÷ Start)^(1 ÷ Years) − 1

End = ending equity, Start = starting equity, Years = the length of the test period in years (use calendar days ÷ 365.25, or trading days ÷ 252, and include fractional years). The result is a decimal rate; multiply by 100 for a percentage. CAGR uses only the two endpoints, so it ignores the entire path between them.

CAGR vs Absolute return vs Annualized return

AspectCAGRAbsolute returnAnnualized return
Time-normalisedYes (geometric)NoYes
Uses full pathNo, endpoints onlyNo, endpoints onlyDepends on method
Comparable across periodsYesNoYes
Accounts for compoundingYesNoYes if geometric
Main blind spotIgnores volatility and drawdownIgnores time and riskIgnores volatility and drawdown

Practical example

Illustrative example (Indian market)

Suppose a Nifty swing strategy is backtested from 1 Jan 2021 to 1 Jan 2024, growing a starting capital of ₹5,00,000 into ₹7,35,000. The period is exactly 3 years. CAGR = (7,35,000 ÷ 5,00,000)^(1 ÷ 3) − 1 = (1.47)^(0.3333) − 1 ≈ 1.1371 − 1 = 0.1371, or about 13.7 percent per year. Note that the simple total return was 47 percent over 3 years, which a careless reader might annualise as 47 ÷ 3 ≈ 15.7 percent; the geometric CAGR of 13.7 percent is the correct, lower figure because it respects compounding.

When comparing a strategy CAGR to “the market” in India, compare against a total-return index (Nifty 50 TRI), not the price index, because the price Nifty excludes dividends and will understate the benchmark by roughly 1 to 1.5 percent a year, quietly flattering your strategy's relative CAGR.

Advantages

  • Puts strategies of different test lengths on one comparable annual scale
  • Correctly reflects compounding and volatility drag, unlike an arithmetic average
  • Simple to compute and universally understood by investors and allocators
  • Directly comparable to a benchmark index CAGR

Limitations

  • Blind to the path: ignores volatility, sequence of returns and every drawdown between endpoints
  • Two strategies with equal CAGR can carry utterly different risk
  • Highly sensitive to the exact start and end dates chosen, especially over short periods
  • A single lucky final month can dominate the endpoint and inflate CAGR
  • Says nothing about capacity, costs or whether the edge persists out-of-sample

Why it matters in practice

  • It is the headline growth number allocators anchor on, so its blind spots must be disclosed alongside it
  • Pairing CAGR with max drawdown (via Calmar) is the minimum honest presentation

Common mistakes

  • Reporting CAGR without a drawdown or risk-adjusted ratio beside it
  • Dividing total return by the number of years instead of taking the geometric root
  • Using whole years when the period includes fractions, which inflates short-period CAGR
  • Cherry-picking start and end dates so the endpoints flatter the rate
  • Comparing a strategy CAGR against the price Nifty rather than the total-return index
  • Trusting a high CAGR from a 2-year backtest as if it were a stable long-run estimate

Professional usage

Professional researchers treat CAGR as a summary statistic, never a verdict. They always report it with a drawdown figure and a risk-adjusted ratio, they compute years to fractional precision, and they check that the CAGR is not an artifact of the exact window by re-running on rolling sub-periods. Allocators know that a smooth 12 percent CAGR and a violent 12 percent CAGR are completely different products, so a serious tear sheet never lets CAGR stand alone.

Key takeaways

  • CAGR is the constant yearly rate that compounds Start into End over the test period
  • It is geometric, so it correctly captures compounding and volatility drag
  • It is blind to the path: equal CAGRs can hide wildly different risk
  • Measure the number of years precisely, including fractions
  • Never report CAGR without a drawdown or risk-adjusted ratio beside it

Frequently asked questions

What is CAGR in a backtest?
CAGR is the single constant annual growth rate that would compound your starting equity into your ending equity over the test period. It is computed as (End ÷ Start) raised to the power (1 ÷ Years) minus one, and it smooths a jagged equity curve into one comparable yearly percentage.
How is CAGR different from total return?
Total return is the raw percentage gain from start to finish regardless of how long it took. CAGR divides that growth across the years geometrically, so a 47 percent total return over 3 years is a CAGR of about 13.7 percent, not 47 percent.
Why use CAGR instead of averaging yearly returns?
Averaging yearly returns arithmetically overstates growth because it ignores compounding and volatility drag. CAGR is a geometric mean, so a plus 50 percent then minus 50 percent pair correctly shows a loss rather than a misleading zero.
What is the biggest weakness of CAGR?
It only sees the first and last equity values, so it is completely blind to the path in between. Two strategies with the same CAGR can have entirely different volatility and drawdowns, which is why CAGR must be paired with a risk measure.
How do I handle a backtest that is not a whole number of years?
Measure the period precisely, using calendar days divided by 365.25 or trading days divided by about 252, and use that fractional value as the exponent's denominator. Rounding 3.25 years down to 3 will inflate the reported CAGR.
Should I compare CAGR against the Nifty price index or the TRI?
Use the Nifty 50 Total Return Index. The price index excludes dividends and understates the market by roughly 1 to 1.5 percent per year, which would unfairly flatter your strategy's relative CAGR.
Can a high CAGR be misleading?
Yes. A high CAGR from a short backtest, a lucky window or a strategy that took enormous drawdowns can look attractive while being fragile. CAGR rewards the endpoint, not the quality or repeatability of the journey.
How does volatility affect CAGR?
Volatility drags CAGR below the arithmetic average return, and the gap widens as volatility rises. Two strategies with the same average return but different volatility will have different CAGRs, with the calmer one compounding to more wealth.
Is CAGR the same as IRR?
No. CAGR assumes a single lump sum from start to end with no interim cash flows. Internal rate of return (IRR) accounts for deposits and withdrawals along the way, so use IRR when capital is added or removed during the period.
Does CAGR account for trading costs?
Only if your equity curve already includes them. CAGR is computed from the net equity endpoints, so if your backtest applied STT, brokerage and slippage, the resulting CAGR is net; if not, it overstates achievable growth.
Why is my CAGR so sensitive to the end date?
Because CAGR depends only on endpoints, a single strong or weak final month can noticeably move it, especially over short periods. Reporting CAGR over several rolling windows reveals how much of it is date luck.
What CAGR is realistic for a retail strategy?
There is no single answer, and any figure presented as achievable would be misleading. Backtested CAGR is a hypothesis about the past under assumptions; live results are typically lower after costs, slippage and regime change.
Can CAGR be negative?
Yes. If the ending equity is below the starting equity, the growth multiple is less than one and the CAGR is negative, representing a steady annual rate of loss over the period.
Should CAGR ever be reported on its own?
No. On its own it hides risk. Serious research always reports it alongside maximum drawdown and a risk-adjusted ratio such as Sharpe or Calmar so the reader sees both reward and the pain that produced it.

Voice search & related questions

Natural-language questions people ask about CAGR (Compound Annual Growth Rate).

What does CAGR mean in simple terms?
It is the steady yearly growth rate that would turn your starting money into your ending money, ignoring the ups and downs in between.
How do I calculate CAGR?
Divide the ending value by the starting value, raise it to one over the number of years, then subtract one.
Is CAGR the same as average yearly return?
No. CAGR is a geometric average that respects compounding, so it is usually lower than a simple arithmetic average of the yearly returns.
Why should I not trust CAGR alone?
Because it only looks at the start and end points and ignores how deep your drawdowns were, so always read it with a risk number.
Can CAGR be faked with date picking?
Yes, choosing flattering start and end dates can inflate it, so it is safer to check CAGR across several time windows.
Does a higher CAGR always mean a better strategy?
Not necessarily, because a higher CAGR might come with far bigger drawdowns, so you must compare risk-adjusted numbers too.

Sources & references

    Last reviewed 11 July 2026. Educational content only — not investment advice. Markets and rules change; verify current conventions with SEBI, NSE/BSE and your broker.

    ← Previous
    Ulcer Index
    Educational content only — not investment advice. Examples use illustrative numbers and simplified models. Backtested results are hypothetical and trading derivatives involves substantial risk. See our Risk Disclosure and SEBI Disclaimer.